Luminance adaptive biomarker detection in digital pathology images

Liu, Jingxin; Qiu, Guoping; Shen, Linlin

Abstract

Digital pathology is set to revolutionise traditional approaches diagnosing and researching diseases. To realise the full potential of digital pathology, accurate and robust computer techniques for automatically detecting biomarkers play an important role. Traditional methods transform the colour histopathology images into a gray scale image and apply a single threshold to separate positively stained tissues from the background. In this paper, we show that the colour distribution of the positive immunohis-tochemical stains varies with the level of luminance and that a single threshold will be impossible to separate positively stained tissues from other tissues, regardless how the colour pixels are transformed. Based on this, we propose two novel luminance adaptive biomarker detection methods. We present experimental results to show that the luminance adaptive approach significantly improves biomarker detection accuracy and that random forest based techniques have the best performances.

Citation

Liu, J., Qiu, G., & Shen, L. (in press). Luminance adaptive biomarker detection in digital pathology images. Procedia Computer Science, 90, https://doi.org/10.1016/j.procs.2016.07.032

• Journal Article Type: Article
• Acceptance Date: Jan 1, 2016
• Online Publication Date: Jul 25, 2016
• Deposit Date: Oct 18, 2017
• Publicly Available Date: Oct 18, 2017
• Journal: Procedia Computer Science
• Print ISSN: 1877-0509
• Electronic ISSN: 18770509
• Publisher: Elsevier
• Peer Reviewed: Peer Reviewed
• Volume: 90
• DOI: https://doi.org/10.1016/j.procs.2016.07.032
• Keywords: Immunohistochemistry; diaminobenzidine; image analysis; luminance; Random Forest
• Public URL: https://nottingham-repository.worktribe.com/output/799627
• Publisher URL: http://www.sciencedirect.com/science/article/pii/S1877050916312121?via%3Dihub
• Contract Date: Oct 18, 2017